This project is a 4-year study to develop and validate innovative analytical uses of multifrequency bioelectrical impedance to quantify total body fat, body water, and to describe levels of regional body fatness in Black and White adults. The study has two aims: 1) To establish new techniques and models for maximum use of the spectrum of multifrequency bioelectrical impedance data in estimating total body fatness, total body water and levels of regional fatness, and 2) To use models of the spectrum of multifrequency bioelectrical impedance data to describe total body fatness, body water and levels of regional fatness among Black and White adults. The rationale for the study is that single frequency (50 MHz) bioelectrical impedance analysis is an imprecise measure of body fatness. Multifrequency BIA also underestimates body fatness in obese people. Multifrequency BIA appears to avoid these problems. However, the system has not been adequately tested in segments of the population at greatest risk for being obese. Body fatness prediction equations has not been not been developed using multifrequency BIA in Black and White adults. A community sample of 480 men and women (plus 48 for a replicability study), ages 18-65 years will be studied. The sample will be equally divided by race and sex. Participants will also be equally divided by BMI levels (< 75th percentile; 25-75th percentile). Total and segmental (right arm, leg, and trunk) body fatness will be estimated using mathematical regression modeling of the multifrequency bioelectrical spectrums of impedance and phase angles. Estimates of body fatness from multifrequency bioelectrical spectrums will be validated and cross-validated against corresponding estimates from DXA, total body water, and extra-cellular fluid volume. Comparison of the models between sex, race, and BMI groups will be made using MANOVA, post hoc tests, and ANCOVA.